Experiments are proposed to test the hypothesis that the site-specific introduction of DNA damage via oligonucleotide-mediated triple helix formation can stimulate homologous recombination in mammalian cells. The broad, long-term objective of these studies is to learn how targeted DNA damage can be used as a tool for the genetic manipulation of mammalian cells for the ultimate purpose of gene therapy. Preliminary work has shown that by linking a triple helix-forming oligonucleotide to a mutagen, the sequence specificity of triplex formation can be imparted to the action of the mutagen, and so DNA damage can be directed to a specific site within mammalian cells. This laboratory has determined the conditions necessary for triple helix-mediated targeted mutagenesis of a selected gene in mammalian cells, and they have shown that a psoralen-linked TFO can stimulate recombination between duplicated genes in an SV40 vector in monkey cells. It is proposed to extend these results with a series of experiments to study the site-specific stimulation of recombination in mammalian cells. Novel SV40-based shuttle vectors will be constructed to investigate parameters that influence induced extrachromosomal recombination. The position of the triplex-targeted damage will be varied, and several types of DNA damage will be tested. Pathways of induced recombination will be examined by using these vectors in repair-deficient human cell lines. Cell lines will be established to contain duplicated thymidine kinase genes flanking a triple helix target site as a model substrate to examine induced recombination at a chromosomal locus. These experiments will serve as a basis for designing additional cell lines to test triplex-induced recombination between a chromosomal site and a homologous donor DNA fragment. This work will lead to experiments to provoke targeted recombination at the endogenous c-myc locus in rodent cells, as a model for gene therapy of a disease-related gene. The possibility of a further enhancement in gene targeting will be tested using a novel strategy in which the TFO is covalently tethered to the donor DNA fragment. With this hybrid molecule, target site recognition can be mediated by triplex formation, thereby positioning the donor fragment for efficient recombination and information transfer. The investigators will build on positive preliminary results to pursue a systematic examination of this approach, using a series of vectors and reporter gene constructs in mammalian cells. This work will provide the foundation for future efforts to correct mutations in disease-related human genes.